13M114MUPS - Applied parallel programming

Course specification
Course title		Applied parallel programming
Acronym		13M114MUPS
Study programme		Electrical Engineering and Computing
Module		Applied Mathematics, Audio and Video Technologies, Biomedical and Environmental Engineering, Biomedical and Nuclear Engineering, Computer Engineering and Informatics, Electronics and Digital Systems, Energy Efficiency, Information and Communication Technologies, Microwave Engineering, Nanoelectronics and Photonics, Power Systems - Networks and Systems, Power Systems - Renewable Energy Sources, Power Systems - Substations and Power Equipment, Signals and Systems, Software Engineering
Type of study		master academic studies
Lecturer (for classes)		professor PhD Milo Tomašević professor PhD Marko Mišić
Lecturer/Associate (for practice)		professor PhD Marko Mišić asst. Matija Dodović, electrical and computer engineer
Lecturer/Associate (for OTC)		professor PhD Marko Mišić asst. Matija Dodović, electrical and computer engineer
ESPB		6.0	Status	elective
Condition		none
The goal		Introduction to basic concepts in the architecture and programming of parallel systems. Detailed knowledge of various shared memory multiprocessor architectures and the most important technologies, frameworks and libraries for parallel programming.
The outcome		Student should be able to understand the design decisions in the area of parallel systems, to use simulators for evaluation of these decisions, and to write parallel programs using different parallel programming models.
Contents
Contents of lectures		Introduction. Application, technology, and architecture trends. Parallel programming models (shared memory, message passing, and data parallel). Shared memory multiprocessors: UMA and NUMA systems. Cache coherence problem. Snoopy and directory protocols and systems. Interconnection networks. Programming of general purpose applications on GPUs.
Contents of exercises		Programming models based on threads (OpenMP) and message passing (MPI). Programming of general purpose applications on GPUs using CUDA. Lab excercises, using simulators of parallel systems, homework assignments.
Literature
Parallel Computer Architecture (A HW/SW Approach), David Culler, J. P. Singh, Anoop Gupta, Morgan Kaufmann, 2000. Programming Massively Parallel Processors, Wen-mei Hwu, David Kirk, Izzat El Hajj, Morgan Kaufmann, 2022. The Science of Computing The Art of HPC, volume 1, Victor Eijkhout, Edmond Chow, Robert van de Geijn, 3rd edition, 2023 Parallel Programming in MPI and OpenMP, The Art of HPC, volume 2, Victor Eijkhout, 2nd edition, 2023.
Number of hours per week during the semester/trimester/year
Lectures	Exercises	OTC	Study and Research	Other classes
2	2	1
Methods of teaching		auditory class lessons and exercises with presentations, lab exercises, homeworks
Knowledge score (maximum points 100)
Pre obligations		Points	Final exam	Points
Activites during lectures		0	Test paper	65
Practical lessons		0	Oral examination	0
Projects		35
Colloquia		0
Seminars		0